Professor Daniel Geschwind relates difficulties in studying a complex disorder like autism, where no two autistic individuals present with precisely the same symptoms.
One of the difficulties with studying a very complex disorder, such as autism, is that it is a very heterogeneous disorder. That is, every child who has it looks a little bit different from each other – it is very complex. Now we don’t know that, for example, two children with autism are going to have the same genetic basis, but there are all these components of the disorder, called endophenotypes, that is, things that we can measure, such as language and social behavior, even things that we can measure in brain structure, for example, that may be closer to an underlying biological cause, that is, or a genetic cause, that we can measure in those children, we can measure in the unaffected relatives. Generally these endophenotypes, if they are going to be genetically useful, are seen in a kind of intermediate form in the unaffected children. So that if you take a, let’s say, an autistic child who is not speaking, his siblings may have some form of mild language delay, but still be non-autistic, but we can use that information and measure that to have more power to identify genes.